Learning proposals for probabilistic programs with inference combinators

We develop operators for construction of proposals in probabilistic programs, which we refer to as inference combinators. Inference combinators define a grammar over importance samplers that compose primitive operations such as application of a transition kernel and importance resampling. Proposals in these samplers can be parameterized using neural networks, which in turn can be trained by optimizing variational objectives. The result is a framework for user-programmable variational methods that are correct by construction and can be tailored to specific models. We demonstrate the flexibility of this framework by implementing advanced variational methods based on amortized Gibbs sampling and annealing.

Download paper here

Recommended citation: @InProceedings{pmlr-v161-stites21a, title = {Learning proposals for probabilistic programs with inference combinators}, author = {Stites, Sam and Zimmermann, Heiko and Wu, Hao and Sennesh, Eli and van de Meent, Jan-Willem}, booktitle = {Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence}, pages = {1056–1066}, year = {2021}, editor = {de Campos, Cassio and Maathuis, Marloes H.}, volume = {161}, series = {Proceedings of Machine Learning Research}, month = {27–30 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v161/stites21a/stites21a.pdf}, url = {https://proceedings.mlr.press/v161/stites21a.html}, abstract = {We develop operators for construction of proposals in probabilistic programs, which we refer to as inference combinators. Inference combinators define a grammar over importance samplers that compose primitive operations such as application of a transition kernel and importance resampling. Proposals in these samplers can be parameterized using neural networks, which in turn can be trained by optimizing variational objectives. The result is a framework for user-programmable variational methods that are correct by construction and can be tailored to specific models. We demonstrate the flexibility of this framework by implementing advanced variational methods based on amortized Gibbs sampling and annealing.} }